|
| 1 | +.. _rm-migration-scenarios: |
| 2 | + |
| 3 | +================================= |
| 4 | +Migration Scenarios and Use Cases |
| 5 | +================================= |
| 6 | + |
| 7 | +.. contents:: On this page |
| 8 | + :local: |
| 9 | + :backlinks: none |
| 10 | + :depth: 1 |
| 11 | + :class: singlecol |
| 12 | + |
| 13 | +You can use Relational Migrator to migrate one legacy application at a |
| 14 | +time to MongoDB. Depending on your application's needs, you can migrate |
| 15 | +either with or without downtime. |
| 16 | + |
| 17 | +Migrate during Scheduled Downtime |
| 18 | +--------------------------------- |
| 19 | + |
| 20 | +If your application does not require constant uptime, such as in the |
| 21 | +case of an internal application, you can have the migration occur during |
| 22 | +scheduled downtime. Migrating during downtime is the easiest way to |
| 23 | +configure and complete the migration process. |
| 24 | + |
| 25 | +When you start the migration process, Relational Migrator takes a |
| 26 | +snapshot of the data on the legacy application and migrates the data |
| 27 | +from that point in time. During the migration process, your legacy |
| 28 | +application can remain online to support incoming reads, but cannot |
| 29 | +support incoming writes. The length of the migration process depends on |
| 30 | +several factors, such as the amount of data being migrated and your |
| 31 | +connection speed. |
| 32 | + |
| 33 | +After the migration completes, take your legacy application offline and |
| 34 | +launch your new application. |
| 35 | + |
| 36 | +Migrate without Downtime |
| 37 | +------------------------ |
| 38 | + |
| 39 | +If your application must remain online to support reads and writes |
| 40 | +during the migration process, you can use Change Data Capture (CDC) to |
| 41 | +migrate data without downtime. This process is called a **continuous |
| 42 | +sync migration**. |
| 43 | + |
| 44 | +When you start a continuous sync migration, Relational Migrator takes a |
| 45 | +snapshot of the data on the legacy application and tracks database |
| 46 | +updates in near real time. During the migration process, your source and |
| 47 | +destination databases remain in sync. |
| 48 | + |
| 49 | +After the continuous sync migration completes, you can transfer users to |
| 50 | +the new application and shut down the legacy application. |
| 51 | + |
| 52 | +To learn more about sync jobs, see :ref:`rm-sync-jobs`. |
| 53 | + |
| 54 | +Currently Unsupported Migration Scenarios |
| 55 | +----------------------------------------- |
| 56 | + |
| 57 | +Relational Migrator currently does not support the following migration |
| 58 | +scenarios: |
| 59 | + |
| 60 | +- Migrating multiple applications through a phased migration without |
| 61 | + downtime. In this scenario, multiple legacy applications are |
| 62 | + decommissioned and migrated to the new database while the original |
| 63 | + database still supports writes. The migration uses :abbr:`CDC (Change |
| 64 | + Data Capture)` to keep the legacy application in sync with the |
| 65 | + destination database. |
| 66 | + |
| 67 | +- Migrating an operational data store. Depending on your workload, this |
| 68 | + scenario could involve :abbr:`CDC (Change Data Capture)` running |
| 69 | + indefinitely without completing. |
| 70 | + |
| 71 | +The preceding scenarios require long-running :abbr:`CDC (Change Data |
| 72 | +Capture)`, or in some cases involve :abbr:`CDC (Change Data Capture)` |
| 73 | +running indefinitely. Therefore, these scenarios are not recommended |
| 74 | +with Relational Migrator. |
| 75 | + |
| 76 | +Relational Migrator will support more migration scenarios and use cases |
| 77 | +in future releases. |
0 commit comments